[Corpora-List] Phd fellowship on Knowledge extraction from scientific publications in the life science domain exploiting existing knowledge repositories - FBK, Trento (Italy)

Alberto Lavelli lavelli at fbk.eu
Tue May 30 08:42:39 CEST 2017

The HLT-NLP research unit at Fondazione Bruno Kessler in Trento (Italy) is soliciting expressions of interest for an upcoming PhD fellowship. The fellowship is funded by the MelanoBase project, whose main aim is large-scale automatic extraction of actionable information from the biomedical literature and its integration with existing structured knowledge (life science databases). The full description of the project is available here: http://hlt-nlp.fbk.eu/projects/melanobase

TITLE Knowledge extraction from scientific publications in the life science domain exploiting existing knowledge repositories

DESCRIPTION The availability of large amount of scholarly results, mainly in the form of scientific publications, combined with recent advances in machine learning (ML) and natural language processing (NLP) techniques, brings closer the possibility of automated knowledge extraction from the literature. In several scientific fields, the fragmentation among subdisciplines, and the increasing amount of published results, make it difficult to exploit such complex knowledge in an effective way. For example, in the life sciences it is estimated that on average about two papers per minute are published. Unstructured information presented in textual format in the literature could be made more valuable if it was organized, integrated, and linked into structured formats, amenable of automated processing, such as databases. The benefits of such resources could potentially extend beyond research settings and reach into clinical care (evidence-based medicine, personalized medicine). In this context, cutting edge NLP and ML techniques may help to extract and organize massive amounts of information from the biomedical literature, in particular in the life science domain because of the availability of manually curated resources which can be used as training material for ML approaches. To this aim, it is of crucial importance the development of tools that can extract key pieces of knowledge from scientific publications and connect them to existing knowledge repositories, in order to create a knowledge base to be used for improving information access and retrieval and to accelerate future scientific discovery. The most recent ML and NLP techniques (such as word embeddings, neural networks, deep learning) are currently explored in order to pursue this ambitious goal. Strong evaluation methodologies, relying on IR/ML/NLP metrics, are essential for experimentally validating the effectiveness of the proposed approaches. A particular focus of the evaluation is on factorial combinations of alternative solutions in order to estimate individual components contributions and to devise predictive performance models. Human factors play also an important role in evaluation, by considering several factors, such as perceived usefulness, improvements in search task completion, and user dynamics and user signals in interacting with information items. Intermediate results should be validated through participation in shared tasks in national and international evaluation activities. The candidate for the position should have a strong computational background, with some expertise in the above mentioned techniques. An interest in the life sciences would be a plus, but it is not strictly required.

IMPORTANT DATES: the call is not yet open. Potential candidates are strongly invited to contact us ASAP sending an email to lavelli at fbk.eu.

CANDIDATE PROFILE The ideal applicant has: * Master degree in Computer Science, Computational Linguistics or related field; * Excellent programming skills in at least one of the following languages: C++, Java, or Python * Understanding of methods in machine learning and computational linguistics; * Interest or prior experience in the biomedical domain is a plus, but not strictly required; * A good knowledge of written and spoken English.

WORK ENVIRONMENT: The doctoral student will work in the HLT-NLP group at Fondazione Bruno Kessler, Trento, Italy. The group has a long tradition in research on natural language processing. Former students are nowadays employed in leader IT companies around the world.

For further information see: http://hlt-nlp.fbk.eu/projects/melanobase http://www.ontogene.org/current-pr/melanobase

Contacts: Nicola Ferro, Department of Information Engineering, University of Padua, Italy Alberto Lavelli, HLT-NLP, FBK, Trento, Italy (lavelli at fbk.eu) Fabio Rinaldi, Swiss Institute for Bioinformatics and University of Zurich, Switzerland & HLT-NLP, FBK, Trento, Italy -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 5092 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20170530/e45a0704/attachment.txt>

More information about the Corpora mailing list